The broad, long-term objectives of this study are to improve the identification and treatment of alcohol abusing pregnant women. Because the social costs of alcohol abuse during pregnancy are potentially very high, and physician detection rates of alcohol abuse by obstetrics patients have been reported to be less than 25%, development of an accurate screen for problematic pregnant drinkers will enhance their identification and lead to further assessment and treatment if indicated. All pregnant women who present to either the Harvard Faculty .or Resident practices at the Brigham and Women's Hospital in Boston, MA will be screened with the T-ACE, a variant of the CAGE, embedded in a General Health and Habits Survey. Women with positive screens will be randomized to either Assessment and Referral Only or Assessment and Referral with a Brief Intervention. Aims specific to this proposal are to: 1. test the sensitivity and specificity of the T-ACE, reported to detect 7 in 10 inner city, African-American women drinking enough to constitute an embryonic/fetal risk, in a more heterogenous population. The relative utility of the T-ACE will be compared to other measures for identifying problematic alcohol use. The other measures will include the AUDIT, SMAST, the alcohol and drug use modules from the SCID, and collateral reports. 2. compare subsequent drinking and alcohol treatment seeking behavior in women identified to have problematic alcohol use after randomization to Assessment and Referral only or Assessment and Referral with Brief Intervention.